Professional Development

Harvard Researchers Found AI Agents Opened 36% More Pull Requests but Shipped Only 18% More Features
An opinion piece on a new Harvard working paper covering 655 companies: after adopting AI agents, teams opened 36% more pull requests but shipped only about 18% more features, because code review became the bottleneck. A sequel to my August piece on why understanding code still matters.

Learning to Code Matters More in the Age of AI, Not Less
An opinion piece arguing that AI makes learning to code more valuable, not less. Two personal stories about overusing AI, backed by industry data, make the case that when generating code is cheap, reading and understanding it becomes the skill that matters most.

Pros and Cons of Using AI for Testing: A Developer's Journey with CenariusCMS
In this article, we explore how AI assisted in writing tests for his Go-based CMS project, CenariusCMS. While AI isn’t flawless, it helped speed up the testing process and uncovered edge cases that hadn't been considered. The article delves into the benefits and challenges of using AI for testing, highlighting its ability to assist without replacing human intuition.